The ABINIT Users Mailing List ( CLOSED )

["Fernando D. Vila" <fer@tiziano.phys.washington.edu>]

Hi..

  I was going to mention the same points. The distance used for the 
calculation is likely to cause very big convergence problems. Your 
suggestion to use several steps and to use the previous results as an 
initial guess is how I usually proceed.

  The only problem I see is that the result will be completely unphysical: 
A multiconfigurational method is needed to correctly describe bond 
breaking.

Regards, Fer.

On Wed, 3 Aug 2005, Masayoshi Mikami wrote:

> Hi, "Nee how,"
>
> I do not know how it went bad, but my suggestion is :
> System which has peculiar (non-classical) chemical bond
> may have such difficulty with SCF convergence.
>
> In your case, C-O bond length is 3.85 x 2 = 7.7 bohr
> = 4.07 angstrom, which is much longer than typical
> C-O bond length in CO (about 1.2 angstrom ? Kindly check it).
> So, initial DEN/WFK may be carefully prepared, e.g.,
> starting the optimized structure, then running a series of
> SCF calculation by increasing the C-O bond gradually (carefully!)
> wherein the initial DEN/WFK for the n-th SCF calculation
> can be obtained from the (n-1)-th SCF calculation result.
>
> In passing, its spin state should also be carefully examined...
> (Please take a look into the Test_*, some of which may be
> revised to be useful for your case...)
>
> Good luck,
> Masayoshi
>
> On 2005/08/03, at 11:11, occupierliu@yahoo.com.cn wrote:
>
> > Dear abinit user:
> >     When I try to calculate the total energy of two atoms at ground state, 
> > the abinis version 4.4.3 break off undesignedly. The error message in the 
> > log file is:
> >   vtowfk : BUG   For istwf_k=2, observed the following element of evec :
> >     3    1             nan             nan
> >   with a non-negligible imaginary part.
> >   Action : contact ABINIT group.
> >
> >  Delivered    6 WARNINGs and   0 COMMENTs to log file.
> >
> >  leave_new : decision taken to exit ...
> >
> > with warnings like:
> >  scfcge: WARNING -
> >   Potential-based CG line minimization has trouble to converge.
> >   The algorithm is restarted with more secure parameters.
> >
> > I've changed the iscf from default 5 to 3, but it didn't work.
> > Do I need to change some parameters? I appreciate any advice.
> >
> >
> >
> > Input file:
> > # C-O potential energy calculation
> > #
> >
> > #Definition of the unit cell
> > acell: 40 34 34   # The keyword "acell" refers to the
> >                   # lengths of the primitive vectors (in Bohr)
> >
> > #Definition of the atom types
> > ntypat 2          # There is only one type of atom
> > znucl 8 6         # The keyword "znucl" refers to the atomic number of the
> >                   # possible type(s) of atom. The pseudopotential(s)
> >                   # mentioned in the "files" file must correspond
> >                   # to the type(s) of atom. Here, the only type is 
> > Hydrogen.
> > nband 10
> > getwfk -1
> >
> > #Definition of the atoms
> > natom 2           # There are two atoms
> > typat 1 2         # They both are of type 1, that is, Hydrogen
> > xcart
> >   -3.85 20.0 20.0
> >    3.85 20.0 20.0
> > #Definition of the planewave basis set
> > ecut 30.0         # Maximal plane-wave kinetic energy cut-off, in Hartree
> >
> > #Definition of the k-point grid
> > nkpt 1            # Only one k point is needed for isolated system,
> >                   # taken by default to be 0.0 0.0 0.0
> >
> > #Definition of the SCF procedure
> > #iscf  3
> > nstep 1000         # Maximal number of SCF cycles
> > toldfe 1.0d-12   # Will stop when, twice in a row, the difference
> >                   # between two consecutive evaluations of total energy
> >                   # differ by less than toldfe (in Hartree)
> > diemac 1.0        # Although this is not mandatory, it is worth to
> >                   # precondition the SCF cycle. The model dielectric
> > diemix 0.5        # function used as the standard preconditioner
> >                   # is described in the "dielng" input variable section.
> >                   # Here, we follow the prescriptions for molecules
> >                   # in a big box


Ubi dubium ibi libertas.
*******************************************************************************
Fernando D. Vila                Voice    (206)543-9697
Department of Physics           Fax      (206)685-0635
University of Washington        E-mail   fdv@u.washington.edu
Seattle, WA 98195, USA          WWW      http://faculty.washington.edu/fdv
*******************************************************************************